NERC Lessons Learned Summary December 2014

NERC lessons learned published in December 2014 Three NERC lessons learned (LL) were published in December 2014 LL Control System Network Switch Failure LL Bus Differential Power Supply Failure LL Loss of Multiple Generators Due to Control Air Problems

Control System Network Switch Failure A partial failure of one of the redundant core switches on a control system mesh network caused a data communication failure. This resulted in two generating units tripping simultaneously, losing a total net output of 1130 MW. The partial failure of the primary core switch kept the switches ports open for traffic The secondary switch detected the partial failure and opened its ports for communication This caused the network to loop, generating a data storm This blocked the communication between the boiler controls and the burner management control processors on both units which eventually tripped the units The “loss of fault tolerant burner control system communications processor (BCS CP)” was the immediate cause of the unit trips

Control System Network Switch Failure Lessons Learned: Redundant devices are implemented to increase reliability, but implementation of such devices may introduce unanticipated failure scenarios if not fully tested Whenever practical, consider a reliable external monitor that can provide diagnostics and alarming to reduce the risk of an undetected failure Consideration should be given to testing and verification of the network topology and fail-over function

Bus Differential Power Supply Failure A microprocessor bus differential relay scheme hardware failure initiated a double bus trip on the BES, resulting in a loss of 58,000 customers The “A” phase differential relay power supply capacitor began degrading and caused the analog to digital converter to provide erroneous current and voltage This resulted in an “A” phase differential trip for bus-1 and bus-2; the voltage supervision was not effective since the degraded capacitor also provided erroneous voltage values used in the differential element supervision This version of the relay does not monitor the power supply internal logic voltages, so the relay did not take itself out of service The double bus trip resulted in a sustained loss of over 58,000 customers

Bus Differential Power Supply Failure Lessons Learned For high-impact schemes, the supervision should be independent of the tripping device In this case, the mode of failure affected the supervising element along with the tripping element (current) being measured The design of this scheme should have involved increased security since one scheme protects both busses Relay manufacturers should ensure there is sufficient device self- monitoring to allow the device to be disabled prior to causing an unwanted trip Manufacturers must communicate the risks clearly to the owners and immediately when the problem is discovered

Loss of Multiple Generators Due to Control Air Problems On April 17, 2013, NERC published Lesson Learned #LL That lesson learned was focused on the inadequate separation of the critical ac power supply of a generating plant’s air system. When the air system was lost, it contributed to the loss of two large generators, creating a system disturbance. Since LL was published, the NERC Event Analysis (EA) staff has identified other instances of generating plants tripping multiple generators due to the loss of station air when the station’s air systems are tied together.

Loss of Multiple Generators Due to Control Air Problems In one example, power plant “A” was equipped with two air headers, which are normally separated. The plant experienced a problem with an air compressor that was serving one of the headers and plant technicians took that compressor out of service for repair. While this compressor was out of service, the air headers were tied together per plant operating procedure. During the time the air headers were tied together, a valve on one of the headers failed, which degraded the station air supply for the plant and resulted in multiple generating units tripping In another example, power plant “B” with a common air header to both of its units experienced an overheating condition in the compressor room, causing two station air compressors to trip. The resulting drop in air pressure supply required the plant operators to manually trip both units due to load swings In a third example, at power plant “C” a loss of auxiliary cooling water caused all instrument air compressors at a generating station to trip. The resulting loss of instrument air supply caused both of the stations generators to run back and trip

Loss of Multiple Generators Due to Control Air Problems Lessons Learned: Plant personnel should be mindful that, while operating in a single header configuration, a problem with any part of the air system supply or the headers may result in a drop in control air and cause multiple units to trip The amount of time that power plants are in this configuration should be kept to a minimum to reduce the chance of this type of failure occurring When possible, individual air compressors should be fed from separate power supplies The installation of additional air compressors may be required to handle a significant loss of station air Procedures and checklists may also need to be updated or revised to address unforeseen low station air pressure conditions

Link to Lessons Learned Directions to Lessons Learned: Go to > “Program Areas & Departments” tab > “Reliability Risk Management” (left side menu) > “Event Analysis” (left side menu) > “Lessons Learned” (left side menu) NERC’s goal with publishing lessons learned is to provide industry with technical and understandable information that assists them with maintaining the reliability of the bulk power system. NERC requests that industry provide input on lessons learned by taking the short survey. The survey link is provided on each Lesson Learned.